For a long time I've had an idea for a strategy and I will share most of it here. I don't feel that there is anything to lose edge wise, because it is not an edge. Most people don't really know what a pivot is and the most challenging part is always the exit. Lets take any strong trend and work backwards. The first support pivot in a strong uptrend, if we were to go one tick below it and you sell short with a stop above the current high, the probability of you getting stopped out is HUGE. With the trend, pivots do all sorts of stuff but to flip a strong trend from up to down, it is never as easy as a few ticks, or points below the low. In addition lets add another thing. All pa traders know the first test of the level is always the high probability play. We are not concerned with retests for the purposes of what im describing. Using this information we know that by buying the correct pivots (or other PA areas based upon tools of your choice), there is a great chance for at minimum a bounce to some level of s/r where newbs who are inadvertently playing countertrend into the first test of a trend level are keeping their stops. If we buy several of these pivots, the chances for a bounce are really high. Employing good management, and proper account size, you could average into a maximum of 2.5% risk with 75%+ accuracy. Notice, i say average into a maximum risk, not averaging down (unknown risk) ... before anyone says anything about blowing up. So, whats left is logistics: 1) How many levels do you average into? 2) Where do you place the price action based stop after X amount of levels? 3) Whats a good PM to allow you to scaleout some but leave the rest to do what trends do (make HHs LLs). I will furthermore add that even though pivots sometimes flip, a trend may still hold if we complete an ABC type retrace. So, is there a way to take advantage of this PA behavior as well? Note, I do not currently do any averaging techniques so Iâm just throwing this around perhaps through a free exchange of ideas, we can get somewhere⦠All the best...
Are more and more people getting interested in trend-following or it is just my idea this is the case? Averaging into trends is the way to make money but also lose fortunes.
There's a a fundamental flaw with your concept. You need strong retracements that end up resuming the trend to make good money. Yet strong retracements might increase chances of trend reversals as the strongest of trends do light pullbacks. The concept of making money with a light position and getting stopped out on a full position is also not very inviting and this is something very possible with your suggestion. I'm not saying you can't make it work, but the above should be taken into consideration, even with good accuracy, you don't really know how much your winning positions will carry in terms of size yet you do know it's full size on the losers this presents a problem when calculating expectancy. Good luck with the project and best of trading to you. NAD
Most people don't understand the dangers of trend following. For example, check out this blogger: http://www.automated-trading-system.com/trade-monthly-basis/ He discusses a system with 67% drawdown while ignoring what this means in reality. Oh boy, imagine you start with 10 Million of client money and you drop to 4 Million. This is catastrophe. Yet, the blogger ignores this, acting at the same time as an expert in the field of trading system development.
The backtest in the article is article is proof of concept. For this purpose the best trading system is a simple one. Simple means easy for the readers to replicate and the complexity of the system doesn't diverge attention from the main idea (which is in this case the effect of using monthly data in a backtest instead of daily data). Same goes for other articles in the blog. Any system published is there with the sole pupose to illustrate an idea. You don't expect the author to publish any trading systems that are actually suitable for trading 10 mln of someone's money, do you?
Moving the stop whenever the size is increased can alleviate this issue. For example, the stop can be moved so that the monetary loss if stopped remains constant; i.e. when a position size is doubled, the stop in price point from the average entry price is halved.
Well by doing this you would be interfering with the one variable that gives this strategy any hope, high accuracy. NAD
lol couldn't be more true. But at the same time, what's a trend and how do you extract a profit from it. NoDoji took the easy way out and said always expect continuation. That advice is good but you don't learn too much from it and it's going to be real hard to stick to it after you first string of losses. I have several ways to answer this question, however the best is a strategy that few people have ever thought of. What does make a trend?? what goes into those MAs?